Online analytical processing (OLAP) is a type of software technology that enables an analyst, an administrator, or an executive to rapidly, consistently, and interactively store and acquire information from multiple perspectives, thereby obtaining a deeper understanding of data. OLAP may be implemented in multiple manners, and MOLAP based on multidimensional data is a main implementation manner. In MOLAP, multidimensional data used in OLAP is acquired from a database, and the foregoing data is physically stored in a form of a multidimensional array to form a “cube”. The cube is formed by two parts: a model of the cube and data stored in the cube, where the model of the cube refers to a data structure that is used to store data. Picturesquely, assuming that the data is an object, the model of the cube is a container that is used to accommodate the object, and the data structure of the model of the cube is a shape structure of the container. Using a cube to store data can improve query performance of OLAP.
However, when a model of a cube or stored data needs to be updated, in the prior art, the cube is reconstructed by deleting the cube and reconstructing a new cube. For example, when a model of a cube needs to be reconstructed, the old cube is deleted, data that needs to be stored in the cube is acquired again from a database, and the acquired data is stored in a model of a new cube, thereby implementing reconstruction of the cube; or when data in a cube needs to be reconstructed, the old cube is deleted, non-updated data is extracted again from a database, and to-be-updated data and the non-updated data that is extracted again from the database are stored again in a cube, thereby implementing reconstruction of the cube.
However, after data in a database is stored in a cube, the data in the database may be lost due to a reason, and integrity of the data in the cube may be damaged when the cube is reconstructed by using the solution in the prior art and based on the data stored in the database.